Hubble AM5 SOC only reaching 99%

Question

question

bash asked Mar 16, '23

Hubble AM5 SOC only reaching 99%

Dear Victron Community,

I am running the following:

Multiplus II 3000VA and a Hubble AM5 battery (5.12 kWh). System is about 2 months old.

It is possible that I am one of a few people running this configuration and so I am hoping someone can offer some assistance.

My issue:

The battery is only being charged to 99%. I am concerned because as I understand, cell balancing occurs at 100% SOC. The BMS also triggers high voltage alarms on the Cerbo and on 3 occasions has turned off the inverter.

I have had a few email dialogues with the manufacturer. Their responses and suggestions are not very helpful. I was even told that the inverter is to "small" and that it cannot charge the battery quick enough and that I need a 5kW inverter. I told them that is surely not the case.

Is this something to be concerned about?

The battery is working as it should. The load on the inverter is small, between 200 and 450 Watts. And so, in a 2 hour grid failure, the SOC will only drop about 10%. As of yet I have never seen the battery discharged to 50%.

All the data is logged via the Cerbo so data analysis is easy.

Any help is highly appreciated.

My advice to any reading this is, if possible, go with SolarMD as your choice of battery. They are in a class of 1.

SOC

cerbo-1.jpg (51.9 KiB)

battery-voltage-553.jpg (46.0 KiB)

warning-high-battery-voltage-2.jpg (41.9 KiB)

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Modifications Space

Node Red Space

Answer 1 · 2023-03-16T12:25:13Z

groningen answered · Mar 16, 2023 at 12:25 PM

@Bash

I cant help you with any answers, but the CLAIM that the inverter is to small......, well, maybe the technician/dealer/sales should go on training. Not trying to upset anyone, just that that claim is false.

If I have to take a guess, I would assume it is the BMS that gives the data to the Cerbo. My theory is, it is the Hubble BMS and not the Victron gear i.e. that either it is not balancing the cells or one of the cells are off.

Can you see the individual cell voltages? (not sure how Hubble works).

Kind regards

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 2 · 2023-03-16T13:26:15Z

nickdb answered · Mar 16, 2023 at 01:26 PM

The BMS is responsible for the SOC calc and sending charge parameters to the system.

It is not uncommon for a lithium battery to not get to 100%, it may be balancing, but at 99% it very much is already balancing at that level, so the battery either needs time or that is just the way their BMS is.

Not all BMS's will religiously show 100%.

Since the BMS is/was reporting high voltage, it would appear to be very slow at balancing, that tends to go away with time and usually why it is recommended to leave a new battery on "keep charged" for a few days so it can sort itself out.

Hubble have had a few wobbles with their BMS - it is a PACE bms which has a chequered past.

It is nonsense that the inverter is too small.

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 3 · 2023-03-16T17:21:29Z

bash answered · Mar 16, 2023 at 05:21 PM

@Groningen @nickdb

Thank you both for your comments. Yes, suggesting that the inverter is charging to slowly shows the kind of intelligence I was dealing with. It was at this point when I knew I was in for a long ride and that I would not be getting the support I needed.

I am unable to see the individual cell voltages. I can only see the lowest and highest cell voltages reported by the Cerbo. No doubt that the manufacturer can see these paramenters, but with suggestions like the above, it means nothing to them.

Thank you Nickdb for the input. It provides some comfort. Something the manufacturer seemingly cannot do.

I have noticed something rather strange. Below is a image which helps to explain.

I removed the BMS CAN cable for a few seconds, see the break in the SOC indicated with the arrow. After I plugged it back, the BMS reported 100%.

This happened twice as shown. the third instance was when I saw the voltage of the battery drop for no obvious reason. I requested the Cerbo to redetect the battery (green arrow). And again, the BMS reports 100% SOC. But as you can see, for about 30 minutes.

I suppose I can live with the problem. It is just somewhat discomforting. Even more so when the manufacturer is unable to provide basic assistance.

I will gladly make the email dialogue available at request.

The latest suggestion was to chnage the inverter settings to AGM and change the absorption voltage to 53.6. Unplug the BMS CAN cable and allow cell balancing to happen.

cerbogx-comms-break-100.jpg (116.5 KiB)

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 4 · 2023-03-16T19:17:30Z

groningen answered · Mar 16, 2023 at 07:17 PM

@Bash

I just scanned through your reply. I am not sure if you should change it to agm. My settings is on lithium. Granted that the bms should "request" voltage if needed.

@nickdb , any suggestions?

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 5 · 2023-03-18T10:16:49Z

bash answered · Mar 18, 2023 at 10:16 AM

@Groningen

I am very reluctant to follow the advice given by the manufacturer, especially changing to the inverter settings to AGM.

I am busy troubleshooting the issue. Although, I probably will not succeed.

I bypassed the Cloudlink device (RIOT Clouldlink), plugged the BMS comms cable directly into the Battery. The battery then would read 100%, for about 30 minutes. During this time, the Battery voltage would slowly drop. I am not sure why. I am hoping the the BMS is intelligent enough and trying to do cell balancing. Although I would need someone to confirm. Unlikely though, the BMS seems rather crude.

I have added the data below.

Another observation. When plugging the comms cable directly into the battery, the battery then reads as PylonTech on the Cerbo. Hahaha. I read about this before on other forums.

On a more serious note. I would like to advocate that Victron remove Hubble as an approved battery. It would seem the BMS is not well suited or tailored to work with Victron and their tech support is not able to help when it comes to Victron integration.

Can someone give me an email address to whom I can forward my concern.

soc-100-99.jpg (57.7 KiB)

bms-min-max-cell-voltage.jpg (78.8 KiB)

bat-voltage-and-current.jpg (136.7 KiB)

1

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

nickdb ♦♦ commented · Mar 18, 2023 at 10:40 AM

Definitely don't change the parameters.

Maybe double check the system is setup correctly by stepping through their guide.

You also need to make sure the mppt's and inverter have manual settings configured for the battery. Some lazy installers don't change all the defaults and assume the battery sets everything, something which goes out the window when the GX is disconnected.

I would check if there is newer battery firmware.

You can access detailed BMS data if you get the right serial cable and download PBMStools, I have used this with pace BMS's before and it helps give better visibility.

0 ·

Answer 6 · 2023-04-03T08:22:18Z

groningen answered · Apr 03, 2023 at 08:22 AM

@Bash

Just doing a followup. Did you reach 100%?

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 7 · 2023-04-12T07:51:54Z

bash answered · Apr 12, 2023 at 07:51 AM

@Groningen

No, and in fact SOC going lower. 97%. I will provide a schematic of the setup in the coming days. I have an idea of what the problem is. There is a third party device (Hubble clouldlink) which is between the Cerbo and the Battery. The manufacturer said that it is required.

When I bypass said device, the Cerbo picks up the Battery as PylonTech (No issue there), but the battery does not charge.

I will contact Hubble and find out why this is the case. My concern is that they will blame the inverter again or anything else rather than admit their product is defective or needs a patch.

The BMS charges the battery to 55.84 V. This is slightly too high for my liking.

Anyway. For those reading this post. Stay away from Hubble Lithium product. Especially if you intent to integrate with Victron products.

2

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

nickdb ♦♦ commented · Apr 12, 2023 at 07:59 AM

Only old batteries needed cloudlink. Current gen can be directly connected to the GX BMS.can port.

If it presents as a pylon that will limit charge, you need to upgrade the Cerbo to v3-32 beta which changes that and will resolve your issue.

1 ·

bash nickdb ♦♦ commented · Apr 15, 2023 at 08:27 AM

@nickdb

Thank you Nick. I will attempt this as soon as I can.

I am hoping that the battery will enter charging when connected directly to the Cerbo (at v3-32). The battery discharges, and I can see that cell balancing is much better.

0 ·

question